Search

In order to asymptotically track the desired trajectory of nonlinear systems with unidirectional control input under various uncertainties, a modified saturated learning-based control law with adaptive notch filter (ANF) is presented in this paper. The proposed control law ensures the asymptotic convergence of the tracking error by utilizing a saturated function and a repetitive learning-based estimator, which allows the learning of the periodic dynamic behavior of the nonlinear system online to make compensations. Simultaneously, an ANF is integrated with the proposed control law to estimate the frequency of the periodic dynamics online. Differing from the existing methodologies in the literatures, the control law proposed in this paper only requires the unknown dynamics be bounded (the parameters, structure, and period of the unknown dynamics is not required to be known). Finally, applying the proposed control law to the electromagnetic suspension system, simulation, and experimental results are included to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

One of the most efficient means to understand complex data is by visualizing them in two- or three-dimensional space. As meaningful data are likely to be high dimensional, visualizing them requires dimensional reduction algorithms, which objective is to map high-dimensional data into low-dimensional space while preserving some of their underlying structures. For labeled data, their low-dimensional representations should embed their classifiability so that their class-structures become visible. It is also beneficial if an algorithm can classify labeled input while at the same time executes dimensional reduction to visually offer information regarding the data’s structure to give rational behind the classification. However, most of the currently available dimensional reduction methods are not usually equipped with classification features, while most classification algorithm lacks transparencies in rationalizing their decisions. In this paper, the restricted radial basis function networks (rRBF), a recently proposed supervised neural network with low-dimensional internal representation, is utilized for visualizing high-dimensional data while also performing classification. The primary focus of this paper is to empirically explain the classifiability and visual transparency of the rRBF. [ABSTRACT FROM AUTHOR]

The motivation behind this paper is to explore the issue of passivity and passification for delayed neural networks with Markov jump parameters. A state feedback control approach with non-uniform sampling period is considered. The interest of this paper lies in the thought about another basic uniqueness wound up being less conservatism than watched Jensen’s inequality and takes totally the relationship between the terms in the Leibniz-Newton formula inside the arrangement of linear matrix inequalities. By utilizing the Lyapunov-Krasovskii functional strategy, a novel delay-dependent passivity criterion is developed with respect to linear matrix inequalities to guarantee the Markov jump delayed neural frameworks to be passive. Passivity and passification problems are tackled by using mode-dependent non-uniform sampled-data control. Using many examples from the literature, it is exhibited that the proposed stabilization theorem is less direct than past results. Finally, the framework is associated with benchmark issue, exhibiting to gather compelling stability criteria for reasonable issues, using the proposed strategy. [ABSTRACT FROM AUTHOR]

The global synchronization of coupled neural networks with hybrid coupling has been studied in this paper. The hybrid coupling is formed from constant coupling, discrete-delay coupling, and distributed-delay coupling. In this regard, a larger class and more complicated coupled neural networks lead in the synchronization problem procedure. According to the new augmented Lyapunov-Krasovskii functional and the idea of M-segmentation of delay length, a less conservative delay-dependent criterion is obtained and expressed in the form of linear matrix inequalities. In many cases, due to the increasing segmentation number, the delay length M-segmentation method could give an opportunity to the user to find a bigger upper bound of the maximum allowable time delay. The effectiveness of suggested method above is proved by simulating a numerical example on a typical chaotic cellular neural network. The results show that the above-mentioned method is less conservative than the other methods reviewed in this article. [ABSTRACT FROM AUTHOR]

In this paper, we present a new solving approach for a class of multi-leader-follower games. For the problem studied, we firstly propose a neural network model. Then, based on Lyapunov and LaSalle theories, we prove that the trajectory of the neural network model can converge to the equilibrium point, which corresponds to the Nash equilibrium of the problem studied. The numerical results show that the proposed neural network approach is feasible to the problem studied. [ABSTRACT FROM AUTHOR]

The Hopfield network is a form of recurrent artificial neural network. To satisfy demands of artificial neural networks and brain activity, the networks are needed to be modified in different ways. Accordingly, it is the first time that, in our paper, a Hopfield neural network with piecewise constant argument of generalized type and constant delay is considered. To insert both types of the arguments, a multi-compartmental activation function is utilized. For the analysis of the problem, we have applied the results for newly developed differential equations with piecewise constant argument of generalized type beside methods for differential equations and functional differential equations. In the paper, we obtained sufficient conditions for the existence of an equilibrium as well as its global exponential stability. The main instruments of investigation are Lyapunov functionals and linear matrix inequality method. Two examples with simulations are given to illustrate our solutions as well as global exponential stability. [ABSTRACT FROM AUTHOR]

In this paper, a class of high-order Hopfield neural networks with neutral distributed delays is considered. Some sufficient conditions are obtained for the existence, uniqueness and global exponential stability of weighted pseudo-almost automorphic solutions for this class of networks by employing the Banach fixed-point theorem and differential inequality techniques. The results of this paper are completely new. An example is given to show the effectiveness of the proposed method and results. [ABSTRACT FROM AUTHOR]

Visual odometry (VO) is a fundamental and challenging problem in both the computer vision community and the robotics community. VO refers the process of recovering the relative movements of a camera by analyzing the associated image sequence. While VO is generally formulated as descriptors-based feature tracking with outliers rejection and global optimization, these algorithms are not only computationally expensive but also lack robustness. In the paper, a biologically inspired solution to the monocular visual odometry problem was presented, which was named as shunting short-term memory monocular visual odometry. The proposed method is simple and concise in both concept and implementation. To be more specific, it utilizes the shunting short-term memory to represent the key frames and the latest observations and also to adapt to uncertainties and ambiguities. And scan matching scheme is adopted to search the movement that best explained the difference between the latest observation and the key frame. Because of the dynamic properties of the neural network, the proposed method requires neither explicit extraction of features and descriptors, nor outliers detection and bundle optimization. Theoretical analysis in the paper showed that the proposed method has Lyapunov stability and constant computational complexity. The proposed method was also compared with the classical monocular VO algorithm in real indoor environments, and the experimental results proved that the proposed method outperforms the classical method on both effectiveness and robustness. [ABSTRACT FROM AUTHOR]

This paper firstly established the finite element model of steel shell motor, computed modal frequencies on top 6 orders to compare with experimental results and verified the reliability of the finite element model. Then, this paper numerically calculated the electromagnetic force of the motor, inputted it into the verified finite element model and computed the vibration acceleration, velocity, stress and strain of the motor. Constraint and properties of internal materials remained unchanged. Steel shell was replaced by aluminum alloy shell to recompute the vibration acceleration, velocity, stress and strain of the motor and compare with those of steel structure motor. Results showed that the motor of aluminum alloy shell had more obvious vibration characteristics. Finally, this paper put forward neural network model optimized by GA. This model was used to predict the vibration characteristics of the motor of aluminum alloy shell and compare with the real value calculated by finite element, showing good consistency. It indicated that it was feasible to predict the vibration characteristics of the motor based on GA-BPNN model. [ABSTRACT FROM AUTHOR]

Based on the traditional boring bar, a boring bar with friction damper is proposed in the paper. Firstly, the frequency response under different pressures is computed primarily based on the theory, which shows that the proposed boring bar has a certain vibration reduction effect. Secondly, the finite element model of the boring bar is built, and the first 6-order modes are computed, whose results are compared with the experimental value. As a result, the virtual reality of the boring bar is achieved. They are consistent with each other, which show that the finite element model is reliable. Then, the experimental cutting process of the boring bar is researched, which is compared with the simulation model with good coincidence. It is found from the result that the cutting simulation model of the boring bar is effective. Later, based on the verified simulation model, the positive pressure between the friction vibrator and boring bar, cutting speed, feed rate, back cutting depth and other parameters are changed to study the vibration reduction effects of the boring bar with friction damper. PSO (particle swarm optimization)-BP (backpropagation) neural network is then used to optimize the cutting process of the boring bar, and the optimal cutting parameters can be obtained. Finally, these optimized parameters are applied in the boring bar, the vibration reduction effect of the boring bar is verified by means of experiments, and the corresponding result shows that the proposed optimization in this paper is feasible. We can obtain higher quality work piece when we use this boring bar in the actual engineering. [ABSTRACT FROM AUTHOR]

Image contour-based feature extraction method has been applied to some fields of image recognition and virtual reality. However, image contour features are easily susceptible to factors like noise, rotation and thresholds during extraction and processing. To solve the above problem, this paper proposes a contour coding image recognition algorithm based on level set and BP neural network models. Firstly, level set model is employed to extract the contours of images. Secondly, image coding method proposed herein is used to code images horizontally, vertically and obliquely. At last, BP neural network model is trained to recognize the image codes. Validity of the proposed algorithm is verified by using a set of actual engineering part images as well as MPEG and PLANE databases. The results show that the proposed method achieves high recognition rate and requires small samples, which also exhibits good robustness to external disturbances such as noise and image scaling and rotation. [ABSTRACT FROM AUTHOR]

Time series forecasting (TSF) consists on estimating models to predict future values based on previously observed values of time series, and it can be applied to solve many real-world problems. TSF has been traditionally tackled by considering autoregressive neural networks (ARNNs) or recurrent neural networks (RNNs), where hidden nodes are usually configured using additive activation functions, such as sigmoidal functions. ARNNs are based on a short-term memory of the time series in the form of lagged time series values used as inputs, while RNNs include a long-term memory structure. The objective of this paper is twofold. First, it explores the potential of multiplicative nodes for ARNNs, by considering product unit (PU) activation functions, motivated by the fact that PUs are specially useful for modelling highly correlated features, such as the lagged time series values used as inputs for ARNNs. Second, it proposes a new hybrid RNN model based on PUs, by estimating the PU outputs from the combination of a long-term reservoir and the short-term lagged time series values. A complete set of experiments with 29 data sets shows competitive performance for both model proposals, and a set of statistical tests confirms that they achieve the state of the art in TSF, with specially promising results for the proposed hybrid RNN. The experiments in this paper show that the recurrent model is very competitive for relatively large time series, where longer forecast horizons are required, while the autoregressive model is a good selection if the data set is small or if a low computational cost is needed. [ABSTRACT FROM AUTHOR]

In this paper, the stability analysis problem is investigated for a new class of discrete-time singular neural networks with Markovian jump and mixed time-delays. The jumping parameters are generated from a discrete-time homogeneous Markov process, which are governed by a Markov process with discrete and finite state space. The mixed time-delays are composed of discrete and distributed delays. The activation functions are not required to be strictly monotonic and be differentiable. The purpose of this paper is to derive some delay-dependent sufficient conditions such that the singular neural networks to be regular, causal and stochastically stable in the mean square. Finally, numerical examples are also provided to illustrate the effectiveness of the proposed methods. [ABSTRACT FROM AUTHOR]

This paper focuses on the adaptive neural output feedback control of a class of uncertain multi-input-multi-output nonlinear time-delay non-integer-order systems with unmeasured states, unknown control direction, and unknown asymmetric saturation actuator. Thus, the mean value theorem and a Gaussian error function-based continuous differentiable model are used in the paper to describe the unknown asymmetric saturation actuator and to get an affine model in which the control input appears in a linear fashion, respectively. The design of the controller follows a number of steps. Firstly, based on the semigroup property of fractional-order derivative, the system is transformed into a normalized fractional-order system by means of a state transformation in order to facilitate the control design. Then, a simple linear state observer is constructed to estimate the unmeasured states of the transformed system. A neural network is incorporated to approximate the unknown nonlinear functions while a Nussbaum function is used to deal with the unknown control direction. In addition, the strictly positive real condition, the Razumikhin Lemma, the frequency-distributed model, and the Lyapunov method are utilized to derive the parameter adaptive laws and to perform the stability proof. The main advantages of this work are that: (1) it can handle systems with constant, time-varying, and distributed time-varying delays, (2) the considered class of systems is relatively large, (3) the number of adjustable parameters is reduced, (4) the tracking errors converge asymptotically to zero and all signals of the closed-loop system are bounded. Finally, some simulation examples are provided to demonstrate the validity and effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

Semantic mapping is still challenging for household collaborative robots. Deep learning models have proved their capability to extract semantics from the scene and learn robot odometry. For interfacing semantic information with robot odometry, existing approaches extract both semantics and robot odometry separately and then integrate them using fusion techniques. Such approaches face many issues while integration, and the mapping procedure requires a lot of memory and resources to process the information. In an attempt to produce accurate semantic mapping with resource-limited devices, this paper proposes an efficient deep learning-based model to simultaneously estimate robot odometry by using monocular sequence frames and detecting objects in the frames. The proposed model includes two main components: using a YOLOv3 object detector as a backbone and a convolutional long short-term (Conv-LSTM) recurrent neural network to model the changes in camera pose. The unique advantage of the proposed model is that it boycotts the need for data association and the requirement of multi-sensor fusion. We conducted the experiments on a LoCoBot robot in a laboratory environment, attaining satisfactory results with such limited computational resources. Additionally, we tested the proposed method on the Kitti dataset, reaching an average test loss of 15.93 on various sequences. The experiments are documented in this video https://www.youtube.com/watch?v=hnmqwxpaTEw. [ABSTRACT FROM AUTHOR]

Face anti-spoofing is a crucial link to ensure the security of face recognition. This paper proposes a novel face anti-spoofing method, which performs ordinal regression with representative feature strengthening to learn generalized and discriminative representation for the live and spoof face classification. Specifically, we propose a semantic label schema, which encodes the inter-class ordinal relationships among live and various spoof faces into supervision information to supervise deep neural networks to perform ordinal regression. It enables the learned model to finely constrain the relative distances among features of different categories in the feature space according to the ordinal relationships. The representative feature strengthening network is designed to strengthen important features and meanwhile weaken redundant features for the classification decision. It leverages a dual-task architecture that takes the same single image as input and shares representations via feature fusing blocks. The network first fuses hierarchical paired convolutional features of two streams to learn the common concern of the two related tasks and then, aggregates the learned local convolutional features into a global representation by a learnable feature weighting block. The network is trained to minimize the Kullback–Leibler divergence loss in an end-to-end manner supervised by the semantic labels. Extensive intra-dataset and cross-dataset experiments demonstrate that the proposed method outperforms the state-of-the-art approaches on four widely used face anti-spoofing datasets. [ABSTRACT FROM AUTHOR]

Photonics-based neural networks promise to outperform electronic counterparts, accelerating neural network computations while reducing power consumption and footprint. However, these solutions suffer from physical layer constraints arising from the underlying analog photonic hardware, impacting the resolution of computations (in terms of effective number of bits), requiring the use of positive-valued inputs, and imposing limitations in the fan-in and in the size of convolutional kernels. To abstract these constraints, in this paper we introduce the concept of Photonic-Aware Neural Network (PANN) architectures, i.e., deep neural network models aware of the photonic hardware constraints. Then, we devise PANN training schemes resorting to quantization strategies aimed to obtain the required neural network parameters in the fixed-point domain, compliant with the limited resolution of the underlying hardware. We finally carry out extensive simulations exploiting PANNs in image classification tasks on well-known datasets (MNIST, Fashion-MNIST, and Cifar-10) with varying bitwidths (i.e., 2, 4, and 6 bits). We consider two kernel sizes and two pooling schemes for each PANN model, exploiting 2 × 2 and 3 × 3 convolutional kernels, and max and average pooling, the latter more amenable to an optical implementation. 3 × 3 kernels perform better than 2 × 2 counterparts, while max and average pooling provide comparable results, with the latter performing better on MNIST and Cifar-10. The accuracy degradation due to the photonic hardware constraints is quite limited, especially on MNIST and Fashion-MNIST, demonstrating the feasibility of PANN approaches on computer vision tasks. [ABSTRACT FROM AUTHOR]

Reward-modulated Spike-Timing-Dependent Plasticity (R-STDP) is a learning method for Spiking Neural Network (SNN) that makes use of an external learning signal to modulate the synaptic plasticity produced by Spike-Timing-Dependent Plasticity (STDP). Combining the advantages of reinforcement learning and the biological plausibility of STDP, online learning on SNN in real-world scenarios can be applied. This paper presents a fully digital architecture, implemented on an Field-Programmable Gate Array (FPGA), including the R-STDP learning mechanism in a SNN. The hardware results obtained are comparable to the software simulations results using the Brian2 simulator. The maximum error is of 0.083 when a 14-bits fix-point precision is used in realtime. The presented architecture shows an accuracy of 95% when tested in an obstacle avoidance problem on mobile robotics with a minimum use of resources. [ABSTRACT FROM AUTHOR]

Monitoring the perceptual quality of digital images is fundamentally important since digital image transmissions through the Internet continue to increase exponentially. Many automatic image quality evaluation (IQE) metrics have been developed based on image features correlated to image distortions; however, those metrics are only effective on particular image distortion types. In recent years, convolutional neural network (CNNs) have been developed for IQEs. These CNNs first capture image features from distorted images; image qualities are predicted based on the captured image features. Since the CNN weights are randomly initialized and are updated with respect to a loss function, image features which are strongly correlated to image quality are not guaranteed to be captured. In this paper, a hybrid deep neural network (DNN) is proposed by integrating image quality metrics to capture image features which are correlated to image quality; the approach guarantees that significant image features are included to predict image quality. Also, a tree-based classifier namely geometric semantic genetic programming is proposed to perform the overall predictions by incorporating CNN predictions and image features; the approach is simpler than the fully connected network but is able to model the nonlinear image qualities. The performance of the proposed hybrid DNN is evaluated by an image quality database with 3000 distorted images. The mean correlation achieved by the proposed hybrid DNN is 0.57 which is higher than the other tested methods. Experimental results with the t- test, F-test and Tueky's range tests show that the proposed hybrid DNN achieves more accurate image predictions with a 99.9% confidence level, compared to the state-of-the-art IQE metics and the most recently developed CNN for IQEs. [ABSTRACT FROM AUTHOR]

In Mobile Ad hoc Network (MANET), network topology changes as devices/users/nodes move and nodes can serve as a source, destination, or router for the information. One of the challenging tasks in MANET is secure routing because of the change of network topology, open nature of wireless communications and frequent breakage in communication links. Thus, it is critically important to develop a robust and secure routing protocol for sending data from the source to the destination in MANET. In this paper, we investigate black hole attacks where compromised nodes utilize routing protocols for self-exposing which has the shortest route for data forwarding to the destination node. Specifically, the proposed approach enhances the Ad hoc On-Demand Distance Vector routing protocol for combating black hole attacks by leveraging the Firefly Algorithm with Artificial Neural Network. Performance evaluation is carried out using numerical results obtained from several experiments and considering different metrics such as computation overhead, packet delivery rate, throughput and delay. Numerical results show that the proposed approach outperforms the traditional approaches. [ABSTRACT FROM AUTHOR]

In recent years, the Internet of Things (IoT) has received a lot of attention. It has been used in many applications such as the control industry, industrial plants, and medicine. In this regard, a fundamental necessity is to implement security in IoT. To this end, Network intrusion detection systems (NIDSs) have been recently in the detection of network attacks and threats. Currently, these systems use a variety of deep learning (DL) models such as the convolutional neural networks to improve the detection of attacks. However, almost all current DL-based NIDSs are made up of many layers, and therefore, they need a lot of processing resources because of their high number of parameters. On the other hand, due to the lack of processing resources, such inefficient DL models are unusable in IoT devices. This paper presents a very accurate NIDS that is named DFE, and it uses a very lightweight and efficient neural network based on the idea of deep feature extraction. In this model, the input vector of the network is permuted in a 3D space, and its individual values are brought close together. This allows the model to extract highly discriminative features using a small number of layers without the need to use large 2D or 3D convolution filters. As a result, the network can achieve an accurate classification using a significantly small number of required calculations. This makes the DFE ideal for real-time intrusion detection by IoT devices with limited processing capabilities. The efficacy of the DFE has been evaluated using three popular public datasets named UNSW-NB15, CICIDS2017, and KDDCup99, and the results show the superiority of the proposed model over the state-of-the-art algorithms [ABSTRACT FROM AUTHOR]

Coming up with a model which matches biological observations more closely has always been one of the main challenges in the field of artificial neural networks. Lately, an ionic model of reservoir networks containing spiking neurons (ILS-based reservoir network) has been proposed which seems to replicate some of the biological processes we have observed up until now. This paper presents a local learning rule for the ILS-based reservoir inspired by the biological fact that each incoming stimulus causes the formation of new dendritic spines, producing new synapses. This property may result in a higher degree of neuroplasticity, leading to a higher learning capacity. To evaluate the proposed learning rule, which somehow takes the structural plasticity into account, several experiments have been designed that show its stability and its effectiveness on separation property, classification accuracy, and reliability. Several benchmark cases have also been discussed and their results through separation and classification evaluation metrics, demonstrate the superior performance of the ILS-based reservoir network trained by the proposed learning rule compared to the untrained similar network. [ABSTRACT FROM AUTHOR]

Recently, neural architecture search (NAS) has gained a lot of attention as a tool for constructing deep neural networks automatically. NAS methods have successfully found convolutional neural networks (CNNs) that exceed human expert-designed networks on image classification in computer vision. However, there are growing demands for semantic segmentation in several areas including remote sensing image analysis. In this paper, we introduce an evolutionary NAS method for semantic segmentation of high-resolution aerial images. The proposed method leverages the complementary strengths of gene expression programming and cellular encoding to develop an encoding scheme, called symbolic linear generative encoding (SLGE), for evolving cells (directed acyclic graphs) as building-blocks to construct modularized encoder-decoder CNNs via an evolutionary process. SLGE can evolve cells with multi-branch and shortcut connections similar to the Inception-ResNet-like modules which can improve training and inference performance in deep neural networks. In experiments, we demonstrate the effectiveness of the proposed method on the challenging ISPRS Vaihingen, Potsdam and UAVid semantic segmentation benchmarks. Compared with recent state-of-the-art systems, our network, dubbed SLGENet, improves the overall accuracy performance on Vaihingen and Potsdam; and achieves a competitive overall accuracy on UAVid using fewer parameters. Our method achieves promising results in a little time of 2.5 GPU days. [ABSTRACT FROM AUTHOR]

Multimedia IoT (M-IoT) is an emerging type of Internet of things (IoT) relaying multimedia data (images, videos, audio and speech, etc.). The rapid growth of M-IoT devices enables the creation of a massive volume of multimedia data with different characteristics and requirements. With the development of artificial intelligence (AI), AI-based multimedia IoT systems have been recently designed and deployed for various video-based services for contemporary daily life, like video surveillance with high definition (HD) and ultra-high definition (UHD) and mobile multimedia streaming. These new services need higher video quality in order to meet the quality of experience (QoE) required by the users. Versatile video coding (VVC) is the new video coding standard that achieves significant coding efficiency over its predecessor high-efficiency video coding (HEVC). Moreover, VVC can achieve up to 30% BD rate savings compared to HEVC. Inspired by the rapid advancements in deep learning, we propose in this paper a wide-activated squeeze-and-excitation deep convolutional neural network (WSE-DCNN) technique-based video quality enhancement for VVC. Therefore, we replace the conventional in-loop filtering in VVC by the proposed WSE-DCNN model that eliminates the compression artifacts in order to improve visual quality and hence increase the end user QoE. The obtained results prove that the proposed in-loop filtering technique achieves - 2.85 %, - 8.89 %, and - 10.05 % BD rate reduction for luma and both chroma components under random access configuration. Compared to the traditional CNN-based filtering approaches, the proposed WSE-DCNN-based in-loop filtering framework achieves efficient performance in terms of RD cost. [ABSTRACT FROM AUTHOR]

Traffic accidents as a result of driver fatigue and drowsiness have caused many injuries and deaths. Therefore, driver fatigue detection and prediction system have been recognized as important potential research areas to prevent accidents caused by fatigue and drowsiness while driving. In this study, driver fatigue is determined by using behavior-based measurement information. Recent studies show that deep neural network is trending state-of-the-art machine learning approaches. Hence, we propose the deep belief network (DBN) model, a deep learning type, used for classification of the symptoms of fatigue in this study. DBN structure is a kind of neural network. The number of hidden layers within the network and the number of units in each hidden layer play important roles in the design of any neural network. Therefore, the hidden layer and the count of units in the DBN model designed in this paper have been selected as a result of various experiments. A greedy method has been adopted to adjust the structure of the deep belief network. Subsequently, the proposed DBN architecture test on KOU-DFD, YawDD and Nthu-DDD datasets. Comparative and experimental results concluded that the proposed DBN architecture is as robust as the other approaches found in the literature and achieves an accuracy rate of approximately 86%. [ABSTRACT FROM AUTHOR]

As one of the most important research topics in the field of natural language processing, open information extraction has achieved gratifying research findings in recent years. Even if so much effort is put into the work of open information extraction, there are still many shortcomings and great room for improvement in the existing system. The traditional open information extraction task relies heavily on the artificially defined extraction paradigm, and it will produce error accumulation and propagation. The end-to-end model relies on a large number of training data, and it is hard to re-train with the increase of the model. To cope with the difficulty of updating parameters of large neural network models, in this paper, we propose a solution based on the meta-learning framework, we design a neural network-based converter module, which effectively combines the learned model parameters with the new model parameters. Then update the parameters of the original open information extraction model using the parameters calculated by the converter. This can not only avoid the problem of error propagation of traditional models but also effectively deal with the iterative updating of open information extraction models. We employ a large and public Open IE benchmark to demonstrate the performance of our approach. The experimental results show that our model can achieve better performance than existing baselines, and compared with the re-training model, our strategy can not only greatly shorten the update time of the model, but also not lose the performance of the model completely re-trained with all the training data. [ABSTRACT FROM AUTHOR]

Since COVID-19 was declared as a pandemic by World Health Organization in March 2020, 169,682,828 cases have been reported worldwide, with 151,416,570 recovered, and 3,526,647 deaths by May 28, 2021. Oxygen gas cylinders demand is booming globally due to its need for COVID-19's for intensive care. Thus, it is critical for hospitals to know exactly the time of receiving oxygen gas cylinders since this will help in minimizing the fatality rate. In this regards, this paper proposes a Multilayer Perceptron Neural Network-based model to predict the delivery time of oxygen gas cylinders for a real-life logistics data from a company that delivers oxygen gas cylinders to all cities around Saudi Arabia. Besides, Multilayer Perceptron Neural Network is benchmarked to supported vector machine and multiple linear regression. Although all the considered models have the ability to provide accurate prediction results, the findings indicate that the proposed supported vector machine and Multilayer Perceptron Neural Network model provide better prediction results. The analysis was achieved through a methodology to identify factors with the highest impact and build a neural network model. The model was further optimized to identify the best order and select the best subset of input variables. The analysis showed that the neural network model can be used effectively to estimate the delivery time of oxygen gas cylinders. The model illustrated high accuracy of prediction by comparing the predicted values to the actual values. [ABSTRACT FROM AUTHOR]

In this paper, the spatial-temporal dynamics of a delayed reaction-diffusion Hopfield's neural network model (HNNM) under a Neumann boundary condition is considered. Our main concern is the absolute, also called the delay-independent, global or local stability of the trivial and nontrivial steady states of the HNNM with non-monotone activation functions. The dissipativity of the semiflow generated by HNNM is first proved and then the absolute attractivity of both the trivial and the nontrivial steady states is obtained by adopting the idea of connecting the spatial-temporal dynamics of the HNNM with the asymptotic behaviors of a finite-dimensional discrete dynamical system. Specifically, it is shown that the strong attractors of the finite-dimensional discrete dynamical system generated by the nonlinear activation function of the HNNM are indeed the attractors of the corresponding delayed reaction-diffusion HNNM. Numerical simulations are also conducted at last to illustrate the effectiveness of our established results. [ABSTRACT FROM AUTHOR]

This paper presents a novel method for improving the invariance of convolutional neural networks (CNNs) to selected geometric transformations in order to obtain more efficient image classifiers. A common strategy employed to achieve this aim is to train the network using data augmentation. Such a method alone, however, increases the complexity of the neural network model, as any change in the rotation or size of the input image results in the activation of different CNN feature maps. This problem can be resolved by the proposed novel convolutional neural network models with geometric transformations embedded into the network architecture. The evaluation of the proposed CNN model is performed on the image classification task with the use of diverse representative data sets. The CNN models with embedded geometric transformations are compared to those without the transformations, using different data augmentation setups. As the compared approaches use the same amount of memory to store the parameters, the improved classification score means that the proposed architecture is more optimal. [ABSTRACT FROM AUTHOR]

Differing from conventional improvements on backpropagation (BP) neural network, a novel neural network is proposed and investigated in this paper to overcome the BP neural-network weaknesses, which is called the multiple-input feed-forward neural network activated by Chebyshev polynomials of Class 2 (MINN-CP2). In addition, to obtain the optimal number of hidden-layer neurons and the optimal linking weights of the MINN-CP2, the paper develops an algorithm of weights and structure determination (WASD) via cross-validation. Numerical studies show the effectiveness and superior abilities (in terms of approximation and generalization) of the MINN-CP2 equipped with the algorithm of WASD via cross-validation. Moreover, an application to gray image denoising demonstrates the effective implementation and application prospect of the proposed MINN-CP2 equipped with the algorithm of WASD via cross-validation. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive critic neural network controller is designed for a class of discrete-time chaotic system. The critic neural network is used to approximate the long-term function. In contrast with the existing results for discrete-time chaotic systems, in this paper, a near optimal control input can be generated when the long-term function is minimized. It is proven that the tracking error, the adaptation laws and the control input are uniformly bounded. A simulation example is employed to illustrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

In an intelligent transportation system, accurate bus information is vital for passengers to schedule their departure time and make reasonable route choice. In this paper, an improved deep belief network (DBN) is proposed to predict the bus travel time. By using Gaussian–Bernoulli restricted Boltzmann machines to construct a DBN, we update the classical DBN to model continuous data. In addition, a back-propagation (BP) neural network is further applied to improve the performance. Based on the real traffic data collected in Shenyang, China, several experiments are conducted to validate the technique. Comparison with typical forecasting methods such as k-nearest neighbor algorithm (k-NN), artificial neural network (ANN), support vector machine (SVM) and random forests (RFs) shows that the proposed method is applicable to the prediction of bus travel time and works better than traditional methods. [ABSTRACT FROM AUTHOR]

It is a well-known fact that improving forecasting accuracy is an important yet often challenging issue. Extensive research has been conducted using neural networks (NNs) to improve their forecasting accuracy. In general, the inputs to NNs are the auto-regressive (i.e. lagged variables) of one or more time series. In addition, either network outputs or network errors have been used as extra inputs to NNs. In this paper, however, we propose a novel recurrent neural network forecasting model which is called the ridge polynomial neural network with error-output feedbacks (RPNN-EOF). RPNN-EOF has two main types of inputs: auto-regressive and moving-average inputs. The former is represented by the lagged variables of a time series, while the latter is represented by feeding back network error to the input layer. In addition, network output is fed back to the input layer. The proposed recurrent model has the ability to produce more accurate forecasts due to the advantages of learning temporal dependence and the direct modelling of the moving-average component. A comparative analysis of RPNN-EOF with five neural network models was completed using ten time series. Simulation results have shown that RPNN-EOF is the most accurate model among all the compared models with the time series used. This shows that employing auto-regressive and moving-average inputs together helps to produce more accurate forecasts. [ABSTRACT FROM AUTHOR]

The smart grid is a revolutionary, intelligent, next-generation power system. Due to its cyber infrastructure nature, it must be able to accurately and detect potential cyber-attacks and take appropriate actions in a timely manner. This paper creates a new intrusion detection model, which is able to classify the binary-class, triple-class, and multi-class cyber-attacks and power-system incidents. The intrusion detection model is based on a whale optimization algorithm (WOA)-trained artificial neural network (ANN). The WOA is applied to initialize and adjust the weight vector of the ANN to achieve the minimum mean square error. The proposed WOA-ANN model can address the challenges of attacks, failure prediction, and failure detection in a power system. We utilize the Mississippi State University and Oak Ridge National Laboratory databases of power-system attacks to demonstrate the proposed model and show the experimental results. The WOA is able to train the ANN to find the optimal weights. We compare the proposed model with other commonly used classifiers. The comparison results show the superiority of the proposed WOA-ANN model. [ABSTRACT FROM AUTHOR]

As a supplement to the financing of small- and medium-sized enterprises, bank microfinance companies are nonbank financial institutions, and it has played an active role in maintaining the stability of financial markets. However, in the course of the operation of microfinance companies, due to the lack of careful management and risk control, the problem of risk management has become increasingly prominent. The purpose of this paper is to study the microfinance risk based on polygenic artificial neural network, and it provides theory and practice application for risk management of microcredit enterprises. Taking the risk management of China's microfinance companies as the research object, on the basis of previous studies, this paper analyzes the risks of bank microfinance companies. Secondly, the basic theory of neural network model and its transformation function are introduced, and the learning method of neural network. At the same time, the learning algorithm of neural network and its improved algorithm are mainly introduced. It lays a theoretical foundation for the follow-up empirical research. Then, through the empirical study of data-based risk assessment of microcredit of farmers, the sample data are divided into training samples and test samples for comparison. Then, we use MATLAB software to establish a neural network model for farmers' microcredit risk assessment. Finally, in order to make the neural network model of farmers' credit risk assessment better popularize and apply, and effectively reduce the credit risk of farmers' microcredit. The corresponding policy suggestions are put forward, which proves the validity and applicability of the neural network in the field of farmers' microcredit risk assessment. It provides a good basis for rural credit cooperatives to identify the credit risk of farmers. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have obtained high accuracy results for pedestrian re-identification in the past few years. There is always a trade-off between high accuracy and computational time in CNNs. Training CNN is always very difficult as it may take a long time to produce high accuracy results. To overcome this limitation, a novel method parallel stochastic gradient descent (PSGD) is proposed to train a five-hierarchical parallel CNNs that is designed according to pedestrian attributes. Moreover, the momentum correction and adaptive adjustment of learning rate are applied during training process and the time interval for updating parameters is inspected during optimization of parameters selection. The results of this paper prove the effectiveness of proposed PSGD that successfully decreases the training process by five times and surpasses the state-of-the-art methods of pedestrian re-identification in terms of both accuracy and time. The minimum reported running time of the proposed method is 8.7 s which is minimum among all other state-of-the-art methods. These promising results show the efficiency and performance of the proposed model. [ABSTRACT FROM AUTHOR]

Uncertain and imprecise data are inherent to many domains, e.g. casting lightweight components. Fuzzy logic offers a way to handle such data, which makes it possible to create predictive models even with small and imprecise data sets. Modelling of cast components under fatigue load leads to understanding of material behaviour on component level. Such understanding is important for the design for minimum warranty risk and maximum weight reduction of lightweight cast components. This paper contributes with a fuzzy logic-based approach to model fatigue-related mechanical properties of as-cast components, which has not been fully addressed by the current research. Two fuzzy logic models are constructed to map yield strength to the chemical composition and the rate of solidification of castings for two A356 alloys. Artificial neural networks are created for the same data sets and then compared to the fuzzy logic approach. The comparison shows that although the neural networks yield similar prediction accuracy, they are less suitable for the domain because they are opaque models. The prediction errors exhibited by the fuzzy logic models are 3.53% for the model and 3.19% for the second, which is the same error level as reported in related work. An examination of prediction errors indicated that these are affected by parameters of the membership functions of the fuzzy logic model. [ABSTRACT FROM AUTHOR]

In the process of establishing supply chain partnership, partner selection is the key and main step. If the enterprise selects the appropriate supply chain partner, in the material equipment unit price, the material equipment production and the supply ability, the product quality appraisal, the brand meets the demand, the new product development ability, has the bad record, the historical project performance, management system and management level, service level, human resource level, internal information processing level, historical cooperation situation, cooperation will, corporate culture and strategic fit degree, financing support ability, information transmission ability, Enterprise green idea propaganda, product energy consumption or energy consumption ratio, green building project participation in construction, environmental protection and energy-saving product development investment, toxic and harmful raw materials use, green matter The use of flowing green packaging, the use of recycled materials/product recycling, and the treatment of industrial "three wastes" will all produce a series of advantages that cannot be matched by traditional relationships. Then, the whole supply chain competitiveness will be improved. This paper studies the selection and evaluation of the partners in the supply chain environment from the point of view of the current research situation of the cooperative relationship in the supply chain environment at home and abroad. Based on the relevant research results at home and abroad, according to certain principles and methods, combined with experts' evaluation of the future trend of qualitative indicators, a set of evaluation index system of supply chain partners is constructed. The standardized treatment of evaluation index is given. By comparing and analyzing the advantages and disadvantages of the common evaluation methods, the mature BP neural network is applied to the artificial neural network by using the artificial neural network evaluation method. MATLAB software is used to construct neural network. BP neural network is used for training and the trained neural network is used to evaluate an example. The results show that this method can solve the problem of partner selection and evaluation in supply chain environment, and improve the evaluation efficiency. [ABSTRACT FROM AUTHOR]

Ball mills are extensively used in the size reduction process of different ores and minerals. The fill level inside a ball mill is a crucial parameter which needs to be monitored regularly for optimal operation of the ball mill. In this paper, a vibration monitoring-based method is proposed and tested for estimating the fill level inside a laboratory-scale ball mill. A vibration signal is captured from the base of a laboratory-scale ball mill by using a ± 5 g accelerometer. Features are extracted from the vibration signal by using different transforms such as fast Fourier transform, discrete wavelet transform, wavelet packet decomposition, and empirical mode decomposition. These features are given as input to an artificial neural network which is used to predict the percentage fill level inside the ball mill. In this paper, the predicted fill level obtained by using different features are compared. It is found that the predicted fill level due to features obtained after fast Fourier transform outperforms other transforms. [ABSTRACT FROM AUTHOR]

In this paper, a hybrid system for wind power ramp events (WPREs) detection is proposed. The system is based on modeling the detection problem as a binary classification problem from atmospheric reanalysis data inputs. Specifically, a hybrid neuro-evolutionary algorithm is proposed, which combines artificial neural networks such as extreme learning machine (ELM), with evolutionary algorithms to optimize the trained models and carry out a feature selection on the input variables. The phenomenon under study occurs with a low probability, and for this reason the classification problem is quite unbalanced. Therefore, is necessary to resort to techniques focused on providing a balance in the classes, such as the synthetic minority over-sampling technique approach, the model applied in this work. The final model obtained is evaluated by a test set using both ELM and support vector machine algorithms, and its accuracy performance is analyzed. The proposed approach has been tested in a real problem of WPREs detection in three wind farms located in different areas of Spain, in order to see the spatial generalization of the method. [ABSTRACT FROM AUTHOR]

Artificial neural network (ANN) aimed to simulate the behavior of the nervous system as well as the human brain. Neural network models are mathematical computing systems inspired by the biological neural network in which try to constitute animal brains. ANNs recently extended, presented, and applied by many research scholars in the area of geotechnical engineering. After a comprehensive review of the published studies, there is a shortage of classification of study and research regarding systematic literature review about these approaches. A review of the literature reveals that artificial neural networks is well established in modeling retaining walls deflection, excavation, soil behavior, earth retaining structures, site characterization, pile bearing capacity (both skin friction and end-bearing) prediction, settlement of structures, liquefaction assessment, slope stability, landslide susceptibility mapping, and classification of soils. Therefore, the present study aimed to provide a systematic review of methodologies and applications with recent ANN developments in the subject of geotechnical engineering. Regarding this, a major database of the web of science has been selected. Furthermore, meta-analysis and systematic method which called PRISMA has been used. In this regard, the selected papers were classified according to the technique and method used, the year of publication, the authors, journals and conference names, research objectives, results and findings, and lastly solution and modeling. The outcome of the presented review will contribute to the knowledge of civil and/or geotechnical designers/practitioners in managing information in order to solve most types of geotechnical engineering problems. The methods discussed here help the geotechnical practitioner to be familiar with the limitations and strengths of ANN compared with alternative conventional mathematical modeling methods. [ABSTRACT FROM AUTHOR]

Sketch-to-photo recognition is an important challenge in face recognition because it requires matching face images in different domains. Even the deep learning, which has recently been deployed in face recognition, is not efficient for face sketch recognition due to the limited sketch datasets. In this paper, we propose a novel face sketch recognition approach based on transfer learning. We design a three-channel convolutional neural network architecture in which the triplet loss is adopted in order to learn discriminative features and reduce intra-class variations. Moreover, we propose a hard triplet sample selection strategy to augment the number of training samples and avoid slow convergence. With the proposed method, facial features from digital photos and from sketches taken from the same person are closer; the opposite occurs if the digital photo and sketch are from different identities. Experimental results on multiple public datasets indicate that the proposed face sketch recognition method outperforms the existing approaches. [ABSTRACT FROM AUTHOR]

In order to quantify the time-varying dependent structure between the assets and forecast the portfolio risk accurately, the difference in the preferences for asset risk is taken into consideration in this paper. It is assumed that the new interest rate of asset return sequence is subject to the standard t distribution. A kind of artificial neural network economic forecasting model is put forward. The two-step state transition diagram estimation method for the economic forecasting is deduced, and the forecasting method for the profile risk is obtained. Finally, Shanghai Securities Composite Index and Standard & Poor's 500 Index are selected to verify the feasibility and superiority of the model and method put forward in this paper. At the same time, the model can accurately quantify the time-varying dependent structural characteristics of the two indices after the subprime mortgage crisis. [ABSTRACT FROM AUTHOR]

In order to improve the prediction accuracy and modeling speed of railway freight volume, this paper combines the cargo floating price prediction model with the neural network algorithm (hereinafter referred to as NNA) to establish a prediction model. The railway cargo floating price based on neural network (hereinafter referred to as PMBCFP) relies on the random forest regression algorithm (hereinafter referred to as RFRA). Through the neural network operator in the prediction model of the floating price of goods, the randomness of the original sequence is weakened, and the implicit rules in the series are mined. The characteristics of neural networks are computationally very simple. In addition, a random forest regression algorithm is applied to the optimization, RFRA's choice. Case studies of China's rail freight volume show that RMSE and other indicators are faster. The MAE, MPE and Tell inequality coefficients obtained from this model were 0.0628, 0.0523, 0.0162 and 0.0107, respectively. This model has good prediction results. The time to search for the best parameters of RFRA using the NNA algorithm is 55.656 s, which is 10.462 s less than the time required for traditional cross-validation methods. Therefore, it is suitable for short-term forecasting of railway freight volume. [ABSTRACT FROM AUTHOR]

Active Learning Method (ALM) is one of the powerful tools in soft computing and it is inspired by the human brain capabilities in approaching complicated problems. ALM, which is in essence an adaptive fuzzy learning algorithm, tries to model a Multi-Input Single-Output system with several single-input single-output subsystems. Each of these subsystems is then modeled by an ink drop spread (IDS) plane. IDS operator, which is the main processing engine of ALM, extracts two kinds of informative features, Narrow Path and Spread, from each IDS plane without complicated computations. These features from all IDS planes are then aggregated in the inference engine. Despite the great performance of ALM in different applications, an efficient hardware implementation has remained a challenge, which is mainly due to considerably high memory requirement of IDS operation. In this paper, in a novel approach to IDS operation, we propose an abstract representation of the IDS planes which minimizes the memory requirement and the computational cost, and consequently, benefits the hardware implementation in terms of area and speed. The proposed approach is fully compatible with memristor-crossbar implementation with an adaptive learning capability. Simpler learning algorithm and higher speed make our proposed algorithm suitable for applications where real-time process, low-cost and small implementation are of high priority. Applications in the classification of real-world datasets and function approximation are provided to confirm the effectiveness of the algorithm. Eventually, the paper concludes that the proposed computing structure provides a synergy between artificial neural networks and fuzzy domains. [ABSTRACT FROM AUTHOR]

In the paper, a method of automatic choice of the boundary conditions is presented. This method allows the initial estimation the value of the boundary conditions during the heating operation for the welding process. The solutions described in this paper can successfully support the work of the specialists performing numerical simulations. They also significantly reduce the time required to estimate the boundary conditions parameters consistent with the actual experiment. The accuracy of the solution in the presented method is not dependent on the empirical models of boundary conditions for the heating process. It is also much more universal. Artificial neural networks after the learning process can be used to perform a reverse analysis for the heat treatment process, for example, to determine the parameters of welding process in the production process of steel elements. [ABSTRACT FROM AUTHOR]

Prostate cancer is the fourth most common cancer among all cancers and the second most common cancer in men. The rate of increase in prostate cancer incidence is higher than the overall increase of cancer incidents. 68% of prostate cancer cases are from developed countries. There has been very little research on the most suitable techniques for analysing prostate cancer gene expression datasets to identify those genes that may be most related to prostate cancer. This paper attempts to identify significant (influential) attributes in a well-established prostate cancer gene expression dataset consisting of over 12,533 attributes for 102 samples (50 normal, 52 tumour). Several (7) different statistical and artificial intelligence (AI)-based feature selection methods were paired with four different classifiers, namely ANNs, Naive Bayes, AdaBoost and J48. Prediction experiments are carried using ANNs with unseen sample testing. In our experiments, ANNs outperformed all other approaches for classification with sequential forward feature selection (SFFS), achieving 100% accuracy. Naive Bayes and AdaBoost achieved best accuracy of 96.3 and 93.13% with support vector machine (SVM) attribute selection, whereas J48 could get only 89.21% with SFFS approach. For prediction experiments, ANNs obtained an accuracy of 95.1% with SVM attribute selection (correctly predicting 96 out of 102 samples). Finally, by investigating National Center for Biotechnology Information database it is found that 21 out of 24 attributes (87.5%) that belong to SVM attribute selection have a reference to cancer/tumour, thereby establishing a link between feature selection and biological plausibility. The main contribution of this paper is in identifying the importance of pairing the most appropriate feature selection strategy with the most appropriate classification strategy when dealing with significantly underdetermined data. This paper also emphasizes differences and similarities between the influence of classification and prediction of prostate cancer. There is another new approach we considered while doing the classification and prediction experiments. Apart from using 7 different feature selection approaches, we have derived new set of attributes by adding all attributes (union), selecting common attributes (intersection) and rest of the attributes (not common). [ABSTRACT FROM AUTHOR]

A preface for the August 2008 issue of the journal "Neural Computing & Applications" presented.

Recent publications have proposed the use of tools based on artificial neural networks to infer the cooling capacity of refrigeration compressors from the results of pressure rise tests, which are quick tests used for production quality assurance. However, the typical rigs used in such tests were not designed to evaluate compressor performance, so the uncertainty in the inferred cooling capacity is high. This paper proposes an improved test rig aiming a better correlation of its results with cooling capacity. A committee of multilayer perceptron artificial neural networks was used to make the cooling capacity inferences from the results obtained in the improved test rig. A method that combines bootstrap techniques with Monte Carlo simulations was used to assure the reliability of the results. The average absolute difference observed between the results of the proposed method and the results of traditional tests done in laboratory was 0.35%, with standard deviation of 0.47%. In addition, the average uncertainty of the inferences was 4.3% for the test samples, which is close to the uncertainty of 3.0% observed in traditional tests, both for a coverage probability of 95%. The time required to carry out the proposed test is about 1 min, thus enabling an increase in the sampling of tested compressors with respect to the traditional method used in industry. [ABSTRACT FROM AUTHOR]

The recent developments in unmanned aerial vehicles (UAVs) and indoor hybrid LiFi–WiFi networks (HLWNs) present a significant opportunity for creating low-cost, power-efficient, reliable, flexible, and ad-hoc HLWN-enabled indoor flying networks (IFNs). However, to efficiently operate and practically realize indoor HLWN, a unified physical layer (UniPHY) is indispensable for joint communication (control and data transfer) and sensing (e.g., localization). A UniPHY structure reduces costs and increases overall flexibility for HLWN-based IFNs. While conventional block-based wireless transceivers independently designed for LiFi and WiFi offer mediocre performance for a composite UniPHY waveform, a machine learning-based end-to-end learning framework for UniPHY can improve overall error performance and reduce the complexity of UAV transceiver hardware. Therefore, this paper proposes a novel generic end-to-end learning framework for a UniPHY system that can efficiently enable HLWN. The performance of the proposed learning framework based on deep neural networks (DNNs) and convolutional neural networks (CNNs) is investigated. Additionally, we assess the computational complexity of the proposed DNN and CNN learning frameworks. The results demonstrate that the performance of DNNs and CNNs varies depending on the considered channel model. Specifically, the analysis reveals that CNNs outperform traditional DNNs in WiFi (Rayleigh fading-based) channels. In contrast, traditional DNNs perform better than CNNs in LiFi (additive white Gaussian noise (AWGN)-based) channels. [ABSTRACT FROM AUTHOR]